Identification of Potent Ebola Virus Entry Inhibitors with Suitable Properties for in Vivo Studies

Article abstract of DOI:10.1021/acs.jmedchem.8b00704

Previous studies identified an adamantane dipeptide piperazine 3.47 that inhibits Ebola virus (EBOV) infection by targeting the essential receptor Niemann-Pick C1 (NPC1). The physicochemical properties of 3.47 limit its potential for testing in vivo. Optimization by improving potency, reducing hydrophobicity, and replacing labile moieties identified 3.47 derivatives with improved in vitro ADME properties that are also highly active against EBOV infection, including when tested in the presence of 50% normal human serum (NHS). In addition, 3A4 was identified as the major cytochrome P450 isoform that metabolizes these compounds, and accordingly, mouse microsome stability was significantly improved when tested in the presence of the CYP3A4 inhibitor ritonavir that is approved for clinical use as a booster of anti-HIV drugs. Oral administration of the EBOV inhibitors with ritonavir resulted in a pharmacokinetic profile that supports a b.i.d. dosing regimen for efficacy studies in mice.

Full text of DOI:10.1021/acs.jmedchem.8b00704

Subscriber access provided by NAGOYA UNIV
Article
Identification of potent Ebola virus entry inhibitors
with suitable properties for in-vivo studies
Hu Liu, Ye Tian, Kyungae Lee, Pranav Krishnan, May Kwang-Mei Wang, Sean Whelan,
Emily Mevers, Veronica Soloveva, Benjamin Dedic, Xinyong Liu, and James Cunningham
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00704 • Publication Date (Web): 19 Jun 2018
Downloaded from http://pubs.acs.org on June 19, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
Identification of potent Ebola virus entry inhibitors with suitable
properties for in-vivo studies
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hu Liu,2,# Ye Tian,1,3,# Kyungae Lee,2,§ Pranav Krishnan,1 May KwangꢀMei
Wang,1 Sean Whelan,2 Emily Mevers,4 Veronica Soloveva,5 Benjamin Dedic1,
Xinyong Liu,3 and James M. Cunningham1,2,*
1Division of Hematology, Department of Medicine, Brigham and Women’s Hospital, Boston, MA
02115
2Department of Microbiology and Immunobiology, Harvard Medical School, Boston MA 02115
3Department of Medicinal Chemistry, Key Laboratory of Chemical Biology (Ministry of
Education), School of Pharmaceutical Sciences, Shandong University, 44 West Culture Road,
250012 Jinan, Shandong, PR China
4Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School,
Boston MA 02115
5United States Army Research Institute of Infectious Diseases, Fort Detrick MD 21702
ꢀ 1 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 39
1
2
3
4
5
6
7
8
9
ABSTRACT: Previous studies identified an adamantane dipeptide piperazine 3.47 that inhibits
Ebola virus (EBOV) infection by targeting the essential receptor NiemannꢀPick C1 (NPC1). The
physicochemical properties of 3.47 limit its potential for testing in vivo. Optimization by
improving potency, reducing hydrophobicity, and replacing labile moieties, identified 3.47
derivatives with improved in-vitro ADME properties that are also highly active against EBOV
infection, including when tested in the presence of 50% normal human serum (NHS). In
addition, 3A4 was identified as the major cytochrome P450 isoform that metabolizes these
compounds and accordingly, mouse microsome stability was significantly improved when tested
in the presence of the CYP3A4 inhibitor ritonavir that is approved for clinical use as a booster of
antiꢀHIV drugs. Oral administration of the EBOV inhibitors with ritonavir resulted in a
pharmacokinetic profile that supports a b.i.d. dosing regimen for efficacy studies in mice.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEY WORDS: Ebola virus, antiꢀviral, NiemannꢀPick C1, cLogP, organic synthesis, ADME,
pharmacokinetics
ꢀ 2 ꢀ
ACS Paragon Plus Environment

Page 3 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
INTRODUCTION
7
8
9
Ebola virus (EBOV) is a highly pathogenic enveloped virus1, 2 that caused the epidemic that
began in the western African nation of Guinea in 2014 and spread to Liberia, Sierra Leone and
Nigeria. During this outbreak, more than 28,000 people were infected and more than 11,000
died.3, 4 The unpredictable onset, ease of transmission, rapid progression, and high morbidity and
mortality of EBOV infection have created a high level of public concern that has led to focus on
development of vaccines and antiꢀviral agents.5ꢀ11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Previously, we identified an adamantane piperazine dipeptide 3.47 that inhibited pseudotyped
13
EBOV infection with an IC50 of 13 nM (Figure 1).12,
Using genetic knockouts and
biochemical assays utilizing a photoꢀaffinity labelling analog 3.98, the target of 3.47 was
identified as the lysosome cholesterol transporter NiemannꢀPick C1 (NPC1) expressed in
susceptible cells.12, 14 We found that 3.47 prevents infection by interfering with NPC1 binding12,
13
and activation of the EBOV glycoprotein (GP) that mediates virus penetration into host
cells.12, 14 Recent studies have defined the role of NPC1 in more detail including reports of the
atomic resolution structure of NPC1 bound to EBOV GP.15, 16 Because NPC1 is essential for
EBOV infection, 3.47 and its derivatives represent an attractive class of antiꢀviral agents.
Figure 1. Adamantane dipeptide piperazines 3.47 and 3.98 are potent inhibitors of EBOV
infection.
Although potent, 3.47 is hydrophobic (clogP = 7.2) and poorly soluble (kinetic solubility = 5
ꢀM). In addition, the in-vitro ADME profile of 3.47 is subꢀoptimal. It is rapidly degraded by
hepatic microsomes (Table 1) and contains a methyl ester that contributes to activity but is a
target for hydrolysis. The current study is aimed at identifying potent 3.47 analogs that overcome
ꢀ 3 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 39
1
2
3
4
5
6
these liabilities and thus, are suitable for testing in animal models of EBOV infection. To achieve
this goal, our approach has been to reduce hydrophobicity,17, 18 and to modify metabolic “hot
spots”, such as the methyl ester and Oꢀbenzyl moieties.
7
8
9
Table 1. In vitro ADME properties of 3.47.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
T1/2 (mouse liver
microsome)
T1/2 (human liver
microsome)
Compound Kinetic Solubility
3.47 5.3 ꢀM
1.0 min
1.2 min
ꢀ 4 ꢀ
ACS Paragon Plus Environment

Page 5 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
CHEMISTRY
6
7
8
9
The syntheses of compounds 1a-1, 1a-3, 1a-4, 1a-9 were previously reported.13 Using similar
methods, we synthesized compounds 1a-2, 1a-5 to 8, 1a-10 to 17 as illustrated in Scheme 1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1: General synthetic scheme for compounds 1a-2, 1a-5 to 8, 1a-10 to 17.
O
Br
O
NH
O
N
R2
OH
R2
N
O
N
N
N
5
O
H
H
O
R2
O
R1
O
7
i
1a-2, 1a-5~1a-8, 1a-10 ~ 1a-17
ii
R1
R1
4
O
6
OH
N
v
R1 = H;
H
O
R2 = CN
11-1
NBoc
O
HN
O
8
NH2
N
CN
N
BocN
iii, iv
ii
HN
O
9
10
O
H2N
O
O
O
O
13
O
OH
R3
R3
N
N
R3
OH
iv
H
H
v
O
O
12
11
14
Reagents and conditions: (i) K2CO3, DMF, rt; (ii) NaBH(OAc)3, DME or DCM, 0°C to rt;(iii) K2CO3, H2O2,
DMSO; (iv) 4N HCl in dioxane, rt; (v) HATU or BOP, DIEA, DMF, rt.
Alkylation of the corresponding phenol 4 with the appropriate benzyl bromide gave benzyl
ether 6 in good yields. Reductive amination with 7 (see Scheme 5 for preparation) using sodium
triacetoxyborohydride [NaBH(OAc)3] gave the majority of compounds 1a (Scheme 1), and
either dimethoxy ethane (DME) or dichloromethane (DCM) as solvent worked well in these
cases. For the 4ꢀamide 1a-2, reductive amination of corresponding 6 with Bocꢀprotected
piperazine, followed by standard acidic Boc deprotection, cyano hydrolysis to the primary
amide, and amide coupling provided 1a-2. The acid 11 was synthesized from appropriate acid 12
via amide coupling with t-butyl glycine ester 13, followed by acidic t-butyl removal.
The synthesis of compounds 1a-18 and 1a-19 are depicted in Scheme 2. The crossꢀcoupling
reaction of 2ꢀbromobenzaldehyde 15 and alkene 16 under the catalysis of palladium (II) acetate
[Pd(OAc)2], provided transꢀalkene 17 which, followed by reductive amination, afforded
ꢀ 5 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 39
1
2
3
4
5
6
compound 1a-18. The double bond was then reduced under standard hydrogenation conditions to
provide compound 1a-19.
7
8
Scheme 2: Synthetic scheme for compound 1a-18 and 1a-19.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (i) Pd(OAc)2, Et3N, P(oꢀMeC6H4)3, Toluene, rt; (ii) NaBH(OAc)3, DCM, 0 °C to rt; (iii)
H2, Pd/C, rt.
The synthesis of compound 1a-20 is shown in Scheme 3. The precursor bromide 18 was
prepared similarly as shown in Scheme 5. Suzuki coupling of 18 using palladium(II) acetate
[Pd(OAc)2] and RuPhos as the ligand gave compound 1a-20.
Scheme 3: Synthetic scheme for compound 1a-20.
HO
B
CO2Me
Br
O
N
OH
O
N
CO2Me
N
19-a
N
N
H
N
O
Cl
H
i
O
Cl
18
1a-20
Reagents and conditions: (i) RuPhos, Pd(OAc)2, KF, DME, 65 °C.
The synthesis of diꢀphenyl ether 1a-21 is shown in Scheme 4. Reacting aldehyde 20 with bocꢀ
piperazine 8 under the standard reductive amination conditions gave phenol 21, which was then
subjected to copper (II) acetate [Cu(OAc)2]ꢀpromoted phenyl ether formation to afford
compound 22. Acidic Boc deprotection, followed by amide coupling with acid 11-1 provided 1a-
21.
Scheme 4. Synthetic scheme for compound 1a-21.
ꢀ 6 ꢀ
ACS Paragon Plus Environment

Page 7 of 39
Journal of Medicinal Chemistry
1
2
3
CO2Me
4
5
6
7
8
9
Boc
HO
N
B
OH
OH
HN
O
N
OH
19-b
ii
8
N
Boc
i
Cl
Cl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
20
O
O
N
O
N
iii
11-1, iv
N
N
Boc
CO2Me
N
CO2Me
H
Cl
22
O
Cl
1a-21
Reagents and conditions: (i) NaBH(OAc)3, DCM, 0°C to rt;(ii) Cu(OAc)2, Pyridine; (iii) 4N HCl in dioxane, rt;
(iv) HATU, DIEA, DMF, rt.
Compounds in the small set of focused library (1b-1 to 1b-18) were synthesized as shown in
Scheme 5. Coupling between adamantyl acetic acid 12-1 and glycine amide 23 gave 24 in good
yield. Acidic Boc deprotection gave piperazine 7, which upon treatment with different aldehydes
26 under the standard reductive amination conditions provided the crude products. Each crude
product was further purified by acidꢀfunctionalized silica gel (SilicaꢀPhSO3H) column to afford
1b-1 to 1b-18 in greater than 90% purities.
Scheme 5. Synthetic scheme for compounds 1b-1 to 1b-18.
O
O
O
OH
O
N
O
N
O
12-1
N
N
N
H2N
i
H
O
O
ii
23
24
O
R4
O
NH
26
O
N
R4
N
N
N
iii
N
H
H
O
O
7
1b-1 ~ 1b-18
Reagents and conditions: (i) HATU, DIEA, DMF, rt; (ii) 4N HCl in dioxane, rt; (iii) NaBH(OAc)3, DME or DCM,
0°C to rt.
ꢀ 7 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 39
1
2
3
4
5
6
7
8
9
The synthesis of compounds 2-1 to 2-18 is shown in Scheme 6. Similar to that described in
Scheme 1, reductive amination of 6 with sodium triacetoxyborohydride [Na(OAc)3BH] gave 27.
Standard acidic Boc deprotection, followed by amide coupling with Boc–protected glycine 29
provided 30, which was then converted in two steps to 2-1 to 2-18. Notably, the
trifluoro(adamantanꢀ1ꢀyl)acetic acid 12-2 was obtained by a series of homologation reactions of
commercially available 32.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 6. Synthetic scheme for compounds 2-1 to 2-18.
Reagents and conditions: (i) NaBH(OAc)3, DME or DCM, 0°C to rt; (ii) 4N HCl in dioxane, rt; (iii) HATU, DIEA,
DMF, rt; (iv) LiAlH4, THF, 0°C to 65°C; (v) SOCl2, DCE, 0°C to 70°C; (vi) NaCN, DMSO, 120 °C; (vii) KOH,
EtOH/H2O, 75°C.
The synthesis of compounds 3-1 to 3-7, 3-10 to 3-13 are shown in Scheme 7. Both reactants,
piperazine 37 and the glycine piperazine amide 38, were prepared similarly as compounds 28 or
31 respectively, as described in Scheme 6. Coupling of 37 or 38 with the corresponding acids 11
or 12, using 1ꢀethylꢀ3ꢀ(3ꢀdimethylaminopropyl)carbodiimide (EDCI), afforded compounds 3-1
to 3-7, 3-10 to 3-13. It is noteworthy to point out that, for either reactants 37 or 38 with a free
ꢀ 8 ꢀ
ACS Paragon Plus Environment

Page 9 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
amino NH2 as the R4 group such as in compound 3-6, amide coupling with acids 11 or 12 under
the HATU conditions gave a significant amount of bisꢀacylated side product (with the 2nd
acylation at the aniline NH2), and this issue was alleviated largely using EDCI conditions.
7
8
9
Scheme 7. Synthetic scheme for compounds 3-1 to 3-7, 3-10 to 3-13.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
OH
R3
N
H
O
N
O
11
R4
N
N
R4
R3
N
HN
H
i
O
37
3-1 ~3-7, 3-10 ~ 3-13
O
R3
OH
12
N
R4
N
i
H2N
O
38
Reagents and conditions: (i) EDCI, HOBt, DIEA, DCM, rt.
The synthesis of compounds 3-8, 3-9 and 3-14 are shown in Scheme 8. The intermediates 40
and 43 were prepared similarly as described in Scheme 6, using the standard reductive amination
conditions. Acylation or sulfonylation of 40, or cyclopropane replacement of the bromide in 43
via Suzuki coupling, afforded intermediates 41 and 44 respectively, which were converted to
compounds 3-8, 3-9 and 3-14 via acidic Boc removal, followed by EDCIꢀpromoted amide
coupling.
Scheme 8. Synthetic scheme for compounds 3-8, 3-9 and 3-14.
ꢀ 9 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (i) NaBH(OAc)3, DME or DCM, 0°C to rt; (ii) Cyclopropanesulfonyl chloride,
pyrimidine, DCM, rt (for 3-8); (iii) Cyclopropanecarbonyl chloride, TEA, DCM, rt (for 3-9); (iv) 4N HCl in
dioxane, rt; (v) EDCI, HOBt, DIEA, rt; (vi) K3PO4, (C6H11)3P, Pd(OAc)2, PhMe/H2O), 80°C.
The synthesis of compounds 3-15 to 3-17 are shown in Scheme 9. Acylation of piperazine 8
with cyclic anhydride 45 gave 46 with a high yield. After standard acidic Boc removal, the amide
in 47 was reduced by LiAlD4 to afford deuterated 48. From intermediates 47 or 48, compounds
3-15 to 3-17 were prepared similarly as described in Scheme 8.
Scheme 9. Synthetic scheme for compounds 3-15 to 3-17.
Reagents and conditions: (i) DMAP, DMF, rt; (ii) 4N HCl in dioxane, rt; (iii) EDCI, HOBt, DIEA, DCM, rt; (iv)
LiAlD4, THF, 0°C to 65 °C.
The synthesis of compounds 3-20 and 3-21 are illustrated in Scheme 10. Aldehyde 6-1,
prepared similarly as in Scheme 1, was treated with glycine attached piperazine 49 under
ꢀ 10 ꢀ
ACS Paragon Plus Environment

Page 11 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
standard reductive amination condition to afford 30-1 in good yield. Boc removal, followed by
EDCIꢀpromoted amide coupling, gave compounds 3-20 and 3-21.
7
8
Scheme 10. Synthetic scheme for compounds 3-20 and 3-21.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (i) NaBH(OAc)3, DME or DCM, 0°C to rt; (ii) 4N HCl in dioxane, rt; (iii) EDCI, HOBt,
DIEA, DCM, rt.
The syntheses of compounds 3-23, 3-24 and (-)3-25 are shown in Scheme 11. Compounds 3-
6, 3-10 and (-)3-18 were converted to compounds 3-23, 3-24 and (-)3-25 smoothly upon
treatment with 4ꢀmethylsulfonyl benzaldehyde 26-1 under reductive amination conditions.
Scheme 11. Synthetic scheme for compounds 3-23, 3-24 and (-)3-25.
ꢀ 11 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 39
1
2
3
4
NH2
HN
F
F
F
O
N
F
O
N
SO2Me
5
6
7
8
O
N
N
N
N
H
SO2Me
O
Cl
H
O
Cl
26-1
3-6
3-23
9
NH2
i
HN
F
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
F
F
O
N
F
O
N
SO2Me
N
N
N
N
H
O
Cl
H
O
Cl
(rac)-3-10
(rac)-3-24
NH2
D
D
HN
F
D
D
F
F
F
O
N
O
N
SO2Me
N
O
N
O
N
N
H
Cl
H
Cl
(-)-3-18
(-)-3-25
Reagents and conditions: (i) NaBH(OAc)3, DME or DCM, 0°C to rt.
RESULTS AND DISCUSSION
The initial goal was to identify functional groups to replace the labile 4’ꢀmethyl ester in 3.47
(Table 2). The activities of analogs containing various 4’ꢀsubstitutions were measured by testing
their effects on infection of Vero cells by vesicular stomatitis virus (VSV) particles pseudotyped
with EBOV GP.19
Table 2. Replacement of the methyl ester in compound 3.47.
IC50 (nM)a
No.
R
13
3.47
ꢀ 12 ꢀ
ACS Paragon Plus Environment

Page 13 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4000b
700
1a-1
1a-2
1a-3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2100b
3300b
350
1a-4
1a-5
1a-6
1a-7
1a-8
300
400
270
4400b
220
1a-9
1a-10
aIC50: concentration of compound required to inhibit VSV EBOV infection of Vero cells by 50%.
b: Data from our previous reported work.13
As expected, we found that the 3.47 hydrolysis product 1a-1 containing the 4’ꢀcarboxylic acid
is significantly less active than 3.47. Replacement of the ester with amides or sulfanamides
(compounds 1a-2 to 1a-7) reduced the cLogP, but also the potency. The bioisosteric oxazole 1a-
8 is only moderately active (IC50 = 270 nM), binds tightly to mouse plasma proteins (99.7%) and
is rapidly degraded by mouse microsomes (T1/2 = 1 min). In addition, 1a-8 is also unstable in
mouse plasma, suggesting that one or more of the amide bonds is also sensitive to hydrolysis
(Table 10).
ꢀ 13 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 39
1
2
3
4
5
6
In addition, we also examined the substitution pattern on phenyl ring 1. Previous SAR that led
to the design of the 5ꢀazido crosslinker 3.98 (Figure 1) indicated that substitutions at the 5
position might be beneficial for antiꢀviral activity.12, 13
7
8
9
Table 3. 5-Substituted phenyl ring 1 and methyl sulfone as the replacement for the ester.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
IC50 (nM)a
No.
R
1.4
1a-11
1.1
9
1a-12
1a-13
1a-14
1a-15
O
CO2Me
12
1400
50
5
1a-16
1a-17
aIC50: concentration of compound required to inhibit VSV EBOV infection of Vero cells by 50%.
ꢀ 14 ꢀ
ACS Paragon Plus Environment

Page 15 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Indeed, we found that 5ꢀsubstitution with fluorine (1a-11, IC50 =1.4 nM) or chlorine (1a-12 IC50
= 1.1 nM) improved antiꢀviral activity by almost tenꢀfold with the 5ꢀchloro being slightly
preferred (Table 3). This suggests that the 5ꢀhalo group occupies a small yet crucial hydrophobic
pocket in NPC1. Taking advantage of this new finding, further studies demonstrated that
substitution of methyl sulfone for the carboxymethyl in 1a-11 did not substantially affect the
potency of the product 1a-17 (IC50 = 5 nM). Thus, compared to 3.47, 1a-17 is more potent, less
hydrophobic (cLogP = 5.8) and lacks the unstable carboxymethyl ester.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In addition to the methyl ester, the Oꢀbenzyl moiety is also a potential site for oxidative
metabolism. Taking this into account, we examined the possibilities of replacing this moiety.
Table 4. Substitutions for the O-benzyl moiety.
IC50 (nM)a
No.
R
1500
20
1a-18
1a-19
1a-20
1a-21
250
16
aIC50: concentration of compound required to inhibit VSV EBOV infection of Vero cells by 50%.
Replacement of the Oꢀbenzyl with a more rigid trans-double bond markedly reduced activity
(Table 4, 1a-18, IC50 = 1500 nM). In contrast, 1a-19, which has a saturated carbonꢀcarbon
linkage, is only slightly less potent than 3.47, suggesting that for compounds like 3.47 or 1a-19
ꢀ 15 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 39
1
2
3
4
5
6
7
8
9
in which the linkage between the two phenyl rings is able to rotate, the most active conformation
is the pseudo-cis conformation. Modelling studies indicated that if this is correct, the labile Oꢀ
benzyl linkage might be successfully replaced by a single bond (1a-20) or by an ether linkage
(1a-21). Testing of these compounds revealed that 1a-20 suffered a significant loss of potency
(IC50 = 250 nM). However, the ether linkage was much better tolerated (1a-21, IC50 =16 nM).
This strongly suggests that the etherꢀlinked methyl benzoate in 1a-21 fills the same binding
pocket occupied by the phenyl ring 2 of 3.47.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Encouraged by the finding that the Oꢀbenzyl linkage might not be critical for activity, we
synthesized and tested focused libraries of derivatives in which the Oꢀbenzyl is absent.
Representative results of this line of investigation are summarized in Table 5.
Table 5. Removal of the O-benzyl ring 2.
O
N
R
N
N
H
O
1b
IC50 (nM)a
IC50 (nM)a
No.
3.0
R
No.
R
1300b
1200
1b-10
F
2
5
360
2200
1b-1
1b-11
Cl
1050
2900b
500
1b-2
1b-3
1b-12
1b-13
> 2000
ꢀ 16 ꢀ
ACS Paragon Plus Environment

Page 17 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
1450
670
200
90
1b-4
1b-5
1b-14
1b-15
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
370
950
1b-6
1b-7
1b-16
1350
310
1b-17
1b-18
430
1900
1b-8
1b-9
1000b
aIC50: concentration of compound required to inhibit VSV EBOV infection of Vero cells by 50%.
b: Data from our previous reported work.13
The analysis revealed that replacement of the 2ꢀoxy benzyl moiety with hydrophobic groups
such as halogens and ꢀCF3 (1b-1 to 1b-3) markedly reduced antiꢀviral activity. Triꢀhalo
substitutions on ring 1 such as 1b-4 to 1b-8 were also weakly active. Interestingly, we found that
although the 2ꢀOHꢀcontaining compound 1b-12 (IC50 = 2900 nM) and the 3ꢀbromo 1b-9 (IC50 =
1000 nM) are only moderately active, combining these modifications into one single molecule
1b-13 markedly improved potency (IC50 = 500 nM). Moreover, substitution of chloride for the 5ꢀ
bromo further increased activity (1b-14, IC50 = 200 nM). A number of other polar substitutions
at the 2ꢀposition were also explored (1b-15 to 1b-18). Among them, we found that the 2ꢀNH2ꢀ
ꢀ 17 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 39
1
2
3
4
5
6
7
8
9
substituted 1b-15 is even more potent (IC50 = 90 nM) and less hydrophobic (cLogP = 5.2) than
the 2ꢀOH in 1b-14 (IC50 = 200 nM; cLogP = 5.80). This suggests that a polar interaction between
the 2ꢀOH or 2ꢀNH2 and NPC1 places the 5ꢀhalogen in a favorable position to occupy the crucial
hydrophobic pocket in NPC1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
With the progress achieved in replacing the labile moieties and reducing the hydrophobicity of
the rightꢀhand side, such as 1a-17 and 1b-15, we focused on modification or replacement of the
adamantane moiety (Table 6). Triꢀfluoro substitution of the adamantane (compound 2-3)
maintained activity but failed to improve microsome stability (Table 10). Not surprisingly,
addition of polar moieties to the adamantane, such as compounds 2-6 and 2-7, also reduced antiꢀ
viral activity, confirming that the binding environment around the adamantane is hydrophobic.
Table 6. Modifications of and substitutions for the adamantane.
IC50 (nM)a
No.
2-1
R1
R2
R3
H
Br
500
600
Cl
Br
H
H
Br
Br
2-2
2-3
200
H
H
Br
Br
700
600
2-4
2-5
ꢀ 18 ꢀ
ACS Paragon Plus Environment

Page 19 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
H
Br
500
2-6
2-7
H
CO2Me
> 2000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H
H
Br
520
80
2-8
2-9
CO2Me
H
H
CO2Me
CO2Me
300
50
2-10
2-11
F
CO2Me
18
2-12
Cl
H
CO2Me
CO2Me
15
73
2-13
2-14
H
H
CO2Me
CO2Me
300
2-15
2-16
2900
Cl
CO2Me
160
2-17
H
Br
> 2000
2-18
aIC50: concentration of compound required to inhibit VSV EBOV infection of Vero cells by 50%.
Previously, we found that replacement of the adamantane with an unsubstituted
cyclohexaneacetamide was poorly tolerated (>30ꢀfold loss of activity).13 Gratifyingly, further
studies revealed that functionalized cyclohexanes could be effective replacements. For example,
ꢀ 19 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 39
1
2
3
4
5
6
7
8
9
the methylꢀsubstituted cyclohexane 2-9 has an IC50 of 80 nM. Other preferred cyclohexanes
include the 4,4ꢀdiꢀfluoroꢀcyclohexane 2-11 (IC50 = 50 nM), and the spiroꢀcyclopropaneꢀ
cyclohexane 2-14 (IC50 = 73 nM). By contrast, the diꢀfluoroꢀcyclobutane in compound 2-10 does
not seem to have sufficient size to fill the adamantaneꢀbinding pocket in NPC1. We also tested
whether the cyclopropane ring alone, as in 2-14, was adequate to achieve desired potency, and
found that neither 2-15 (IC50 = 300 nM) nor 2-16 (IC50 = 2900 nM) exhibited significant activity.
Importantly, we confirmed that the halogen substitutions at the 5ꢀposition of phenyl ring 1
further improved the antiꢀEBOV activity of compounds 2-12 (IC50 = 18 nM) and 2-13 (IC50 = 15
nM). Thus, successful replacement of the adamantane resulted in potent inhibitors with
significantly lower cLogPs than 3.47 (2-12, clogP = 5.16 and 2-14, clogP = 5.42).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To identify the most promising compounds for advancement, we evaluated derivatives that
contained the optimized functional groups (Table 7).
Table 7. 2-Amino or -hydroxyl benzyl-piperazinyl-2-(4,4-difluoro-cyclohexyl)-acetamides.
IC50 (nM)a
No.
3-1
R1
R2
OH
Cl
470
2800
3600
3-2
3-3
ꢀ 20 ꢀ
ACS Paragon Plus Environment

Page 21 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
790
180
3-4
3-5
3-6
3-7
F
9
OH
Cl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
F
(rac)
340
1740
2200
3500
100
3-8
3-9
3-10
240
3-11
3-12
2500
ꢀ 21 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 39
1
2
3
4
5
6
7
8
160
3-13
3-14
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4200
>5000
347
3-15
3-16
3-17
94
51
3-18
3-19
F
F
180
H
(+)-enantiomer
aIC50: concentration of compound required to inhibit VSV EBOV infection of Vero cells by 50%.
We found that compound 3-1 containing the diꢀfluoroꢀcyclohexane acetamide replacement for
adamantane and the 2ꢀOH, 5ꢀCl substituted phenyl ring 1 has an IC50 of 470 nM. As previously
found, substitutions at other positions on ring 1 were deleterious (3-2 and 3-3). Importantly,
incorporation of the spiroꢀcyclopropane into 3-1 significantly increased potency (3-5, IC50 = 180
ꢀ 22 ꢀ
ACS Paragon Plus Environment

Page 23 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
nM), indicating that the 4,4ꢀdifluoroꢀcyclohexane spiroꢀcyclopropane combination is the
preferred replacement for the adamantane. As expected, replacement of 2ꢀOH in 3-5 with 2ꢀNH2
was beneficial (3-10, IC50 = 100 nM). Moderate loss of potency was observed when the diꢀ
fluoroꢀsubstituted cyclohexane was replaced by cyclopentane, and further reductions in the ring
size were even more deleterious (3-11 and 3-12). Interestingly, 3-11 and 3-13 possess similar
potencies, with 3-13 being the open ring analog of the former. We also tested 3-16 and 3-17, two
derivatives of 3-6 and 3-10, in which the benzylic position is blocked with deuterides,20 and
found this modification had little impact on potency. Since the cyclopropane has a chiral center,
the racemic 3-17 was separated into the enantiomers, (-)-3-18, and (+)-3-19. We found that (-)-3-
18 (IC50 = 51 nM) is 3ꢀfold more potent than (+)-3-19 (IC50 = 180 nM).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We also evaluated the effect of adding the benzyl moiety to scaffolds such as phenols 3-1 and
3-5, and anilines 3-6, 3-10 and 3-18 (Table 8). We found that the 2ꢀOꢀbenzyl analogs are more
potent than 3.47 (IC50 = 63 nM), with the enantiomer (-)-3-22 being the most active (IC50 =
19nM). Similarly, the 2ꢀNH benzyl derivatives (3-23 to 3-25) showed the same trend, with (-)-3-
25 being the most potent (IC50 = 21 nM). These studies were carried out utilizing the replication
competent VSV vector rVSV14. Compared to pseudotyped VSV, rVSV is slightly less
susceptible to inhibition but significantly more stable when studied in cells cultured in medium
containing high serum to mimic in vivo conditions. Studies using this virus demonstrated that the
potencies of 3-22 and of 3-25 were reduced by only 5ꢀfold when tested in media containing 10%
fetal bovine serum with 50% normal human serum.
Table 8. Evaluation of benzyl containing NPC1 inhibitors.
IC50 (nM)a IC50 (nM)b
No.
R1
R2
64
ꢀ
3.47
ꢀ 23 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 39
1
2
3
4
5
6
7
8
9
1195
163
ꢀ
ꢀ
3-16
3-18
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
38
35
19
ꢀ
ꢀ
3-20
3-21
3-22
93
ꢀ
28
40
3-23
3-24
21
109
3-25
aIC50: concentration of compound required to inhibit rVSV EBOV infection of Vero cells by 50%.
bIC50: concentration of compound required to inhibit VSV EBOV infection of Vero cells by 50% in the
presence of media containing 10% FBS and 50% NHS.
Although the aniline, as in 3-18, 3-23 to 3-25, is considered a structural alert,21 the high fatality
rate and short duration of treatment of EBOV infection may mitigate this concern.
We compared the activities of 3.47, 3-21, 3-23 and 3-24 on the growth of bona fide EBOV
(Table 9). This test measured the effect of these inhibitors on virus production by HeLa cells for
ꢀ 24 ꢀ
ACS Paragon Plus Environment

Page 25 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
two days after infection with EBOV isolates Mayinga (1976), Kikwit (1995) and Makona (2014)
that caused the three largest outbreaks in Africa. We found that 3-23 and 3-24 are substantially
more active than 3.47 against these viruses. Under these conditions, the concentration of these
inhibitors that are required to reduce cell viability by 50% exceeded 10 ꢀM. These findings
demonstrate that the VSV EBOV entry assay is a reasonable predictor of activity against bona
fide EBOV.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 9. Inhibition of EBOV growth in Hela cells.
EC50 (nM)
EC50 (nM)
EC50 (nM)
No.
EBOV Kikwit EBOV Makona EBOV Mayinga
280
210
12
160
140
52
590
420
33
3.47
3-21
3-23
3-24
9
17
35
We examined the impact of replacement of the adamantane and other changes that decreased
hydrophobicity on the in vitro ADME properties of 3-6 (clogP = 2.94), 3-16 (clogP = 2.94), 3-18
(clogP = 2.87), 3-20 (clogP = 4.10), 3-22 (clogP = 3.82) and 3-25 (clogP = 3.24) (Table 10).
Although inhibitors 3-22 and 3-25 are very potent, they are less soluble and bind more
extensively to mouse plasma proteins (plasma protein binding, PPB = 97.6% and 99.8%,
respectively) than smaller compounds such as 3-6, and 3-18 (PPB = 86.9% and 87.9%,
respectively). The stability of each compound in plasma is determined by factors such as the
molecular weight and steric hindrance around the amide bond. Lower molecular weight
compounds, such as 3-6, 3-16 and 3-18, are stable in mouse plasma (> 86% remained after four
hours) regardless of the left hand acetamide structures. However, the larger compound 3-20,
which has the less sterically hindered diꢀfluorocyclohexane acetamide, was unstable in mouse
plasma (30% remained after four hours). By comparison, compound 3-22, in which the αꢀcarbon
of the acetamide is blocked by the methylene of the cyclopropane ring, is more stable (69%
remained after four hours).
The stability of these compounds in mouse liver microsomes (MLM) remained suboptimal and
in particular, deuteride blocking at the benzylic position of 3-6 (T1/2 = 9 min) had minimal impact
ꢀ 25 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 39
1
2
3
4
5
6
7
8
9
on the microsomal stability (3-16, T1/2 = 11 min). To better understand the oxidative metabolic
pathways in mice, we carried out additional studies of mouse liver microsomes that revealed 3A4
as the major P450 cytochrome isoform responsible for metabolism of these compounds (data not
shown). Consistent with this finding, the presence of the CYP3A4 inhibitor ritonavir (3 ꢀM),
which is approved for clinical use as a booster of antiꢀHIV drugs,22 increased the stabilities of 3-
16 (T1/2 > 120 min) and 3-18 (T1/2 =91 min) in mouse liver microsomes by 10ꢀ to 30ꢀfold. The
salutary effect of ritonavir was also conferred on derivatives such as 3-22 and 3-25 that contain
the benzyl moiety. Ritonavir does not inhibit VSV EBOV infection (IC50 > 25 ꢀM), and thus is
not expected to confound the evaluation of lead NPC1 inhibitors as antiꢀEBOV agents. As with
3.47, 3-21, 3-23 and 3-24, the concentration of each of these inhibitors required to reduce cell
viability by 50% exceeded 10 ꢀM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 10. In vitro ADME properties of selected compounds.
No.
Kinetic Mouse
Mouse
plasma
stability
(% at
4 hours)
37
MLM
T1/2
(min)
MLM,
T1/2
(min)
(ritonavir
3 µM)
ꢀ
HLM
T1/2
(min)
solubilit
y (µM)
PPB
(%
bound)
5
ꢀ
99.7
ꢀ
1
1
ꢀ
ꢀ
1a-8
2-3
ꢀ
ꢀ
> 100
93
78
ꢀ
10
4
86.9
ꢀ
87.9
ꢀ
97.6
99.8
94
86
88
30
69
72
9
11
3
ꢀ
ꢀ
ꢀ
27
32
16
ꢀ
ꢀ
1
3-6
> 120
91
> 120
> 120
87
3-16
3-18
3-20
3-22
3-25
2
At this stage, we investigated whether the in vitro optimization of 3.47 along with the
protective effect of ritonavir would translate into an acceptable pharmacokinetic profile in vivo.
Oral pharmacokinetic properties for selected compounds were obtained in mice (Table 11).
Table 11. Oral pharmacokinetic parameters for selected compounds in mice.
Cmax
(ꢀM)
C@12hr AUClast
Cl_obs
(mL/min/Kg)
T
max (hr)
T1/2 (hr)
No.
Dose
(ꢀM)
(ꢀM.hr)
25 mg/kg 0.6
7.2
0.2
19.9
2.3
49.0
3-16
ꢀ 26 ꢀ
ACS Paragon Plus Environment

Page 27 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
25mg/kg/
20mg/kg
3-16/
ritonavir
1.3
1.7
4.0
7.3
12.0
19.5
10.4
7.4
1.7
2.0
4.7
5.7
76.8
2.3
3.2
2.2
1.7
12.3
15.5
11.9
12.5
6
7
8
9
50mg/kg/
50mg/kg
3-18/
ritonavir
120.9
116.4
107.5
50mg/kg/
50mg/kg
3-22/
ritonavir
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50mg/kg/
50mg/kg
3-25/
ritonavir
After administration to mice by gavage, the inhibitor 3-16 was rapidly absorbed and reached a
max of 7.2 ꢀM but was nearly completely cleared in 12 hours. Coꢀadministration of ritonavir (20
C
mpk) increased the AUC and decreased the clearance of 3-16 by nearly 4ꢀfold. As a result, the
plasma concentration of 3-16 after 12 hours increased to 1.7 ꢀM. The pharmacokinetics of a
single 50 mpk dose of 3-18, 3-22 and 3-25 coꢀadministered with ritonavir (50 mpk) was also
studied. The plasma concentrations of 3-22 (4.7 ꢁM) and 3-25 (5.7 ꢁM) after 12 hours were 50ꢀ
fold greater than the in vitro IC50s of these compounds measured against rVSV EBOV infection
in 50% human serum (Table 8). Not surprisingly, when the concentration of ritonavir decreased
after 12 hours, the plasma concentration of the inhibitors declined more rapidly. These findings
suggest that an oral b.i.d dosing regimen that delivers an effective concentration of EBOV
inhibitors 3-22 and 3-25 can be achieved in mice.
CONCLUSIONS
EBOV infection is initiated by fusion of the virus membrane to target cells. Virus membrane
fusion is mediated by a conformational change in the GP that is triggered upon binding to the
lysosome membrane protein NPC1.12 Previous studies identified specific adamantane dipeptide
piperazines that inhibit infection by targeting NPC1 and preventing binding and activation of GP.
We report here the propertyꢀdriven lead optimization of this class of inhibitors. Undesirable
ADME properties in the original lead compound 3.47 included high hydrophobicity (cLogP =
7.2), attributable in large part to the adamantane, and hydrolyzable amide and methyl ester
bonds. SAR efforts identified acceptable replacements for the adamantane with spiroꢀ
cyclopropane diꢀfluoro cyclohexanes and for the ester with methyl sulfone. In addition, we found
that halide substitutions at the 5ꢀposition of phenyl ring 1 increased potency by almost 10ꢀfold.
Those modifications, when combined into a single molecule, resulted in inhibitors, such as 3-
ꢀ 27 ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 39
1
2
3
4
5
6
7
8
9
22 and 3-25, that are more potent and less hydrophobic than 3.47 and are highly active against
EBOV Mayinga (1976), Kikwit (1995) and Makona (2014) that caused the three largest
outbreaks of human infection. Compounds 3-22 and 3-25 are stable in mouse plasma for up to 4
hours. It is important to note that the potencies of 3-22 and of 3-25 were reduced by only 5ꢀfold
when tested in media containing 10% fetal bovine serum and 50% normal human serum.
Furthermore, 3A4 was identified as the major cytochrome P450 isoform that metabolizes these
compounds. Consistent with this finding, oral coꢀadministration of either 3-22 or 3-25 to mice
with ritonavir, a cytochrome P450ꢀ3A4 (CYP3A4) inhibitor with wide clinical use as a booster
for antiꢀvirals, resulted in plasma concentrations of 3-22 and 3-25 at 12 hours post dose that are
50ꢀfold greater than the antiꢀviral IC50 of these compounds measured in high serum. These
findings suggest that oral b.i.d dosing regimens of 3-22 or 3-25 will deliver effective inhibition
of the EBOV receptor function of NPC1 in mice. Selected compounds from this study will be
tested in the mouse model of Ebola infection, and those results will be reported in due course.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemistry. 1H NMR spectra were recorded on a Varian Inova 600 MHz spectrometer with
chemical shifts reported in parts per million (ppm) relative to an internal standard
(trimethylsilane). Coupling constants (J) are reported in hertz (Hz). Standard resolution mass
spectra were obtained on an Agilent 1200 Series HPLC (4.6 x 100 mm, 5 ꢁm Phenomenex C18
reverseꢀphase column) and a 6130 Series mass spectrometer system; all mass spectra were
obtained using electrospray ionization (EI) in positive ion mode. Standard reverseꢀphase HPLC
conditions were as follows: mobile phase A = 0.1% formic acid in water; mobile phase B =
0.1% formic acid in acetonitrile. The same standard reverseꢀphase HPLC conditions were also
used to determine the purity of the synthesized compounds. For singletons, all compounds have
purities > 95%. All libraries compounds (parallel syntheses, 1b-1 to 1b-18) have purities > 90%.
Solvents for synthesis were purchased as anhydrous grade and used without further purification.
Reagents were purchased from commercial sources and used as received.
General synthetic procedure for the preparation of 1a-2, 1a-5 to 8, 1a-10 to 17 (Scheme 1).
To a mixture of the appropriate acid (12, 3.2 mmol, 1 eq), tertꢀbutyl glycinate (13, 3.9 mmol,
1.2 eq), HATU (4.8 mmol, 1.5 eq), under nitrogen was added 6 mL of anhydrous DMF. Under
stirring, N, Nꢀdiisopropylethylamine (0.3 mL) was added and the reaction then stirred for
ꢀ 28 ꢀ
ACS Paragon Plus Environment

Page 29 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
overnight at room temperature. To the mixture was added water (30 mL) and a white wax
precipitate formed. The water was decanted, and the precipitate was washed with water (3 x 30
mL) under sonication, then dried to afford crude product of 14, which was used without further
purification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In a glass vial, the substituted tertꢀbutyl glycinate 14 (2 mmol) was dissolved in 2 mL of
EtOAc and then 2 mL of 4 N HCl in dioxane was added to the solution. The mixture was stirred
at room temperature for overnight and then concentrated by GeneVac. The desired acid 11 was
obtained as a solid, and used without further purifications.
To a mixture of proper phenol aldehyde (4, 3.8 mmol, 1eq) and potassium carbonate (11
mmol, 3 eq) in DMF (5 mL) was added 1ꢀ(bromomethyl)ꢀ4ꢀsubstituted benzene (5, 4.0
mmol,1.05 eq) and the resulting mixture was stirred for 16 h at room temperature. The solvent
was removed under reduced pressure, and the residue was partitioned between H2O (20 mL) and
EtOAc (50 mL). The organic layer was washed with brine, dried over sodium sulfate, and
concentrated under reduced pressure. The residue was purified by flash column chromatography
(gradient: 20ꢀ100% EtOAc/hexanes) to give 6 as a white solid,
In a vial was placed piperazine 7 (0.06ꢀ0.15 mmol, 1.2 to 3 eq, and for the preparation of 7,
please refer to scheme 5) in 2 mL of anhydrous DME or DCM. Aldehyde 6 (0.05 mmol, 1 eq)
o
was then added, followed by NaBH(OAc)3 (0.1ꢀ0.2 mmol, 2 to 4 eq) at 0 C. The reaction was
allowed to warm to rt, and stirred for overnight, and then was dried to afford an oily crude
mixture. To this mixture was added sat. aq.Na2CO3 (5 mL), and the suspension was sonicated for
10 minutes to obtain a sticky brown color wax/oil. The wax/oil was further sonicated with water
(2 x 10 mL). The waxy residue was then dried to afford the crude products. Pure compounds 1a-
2, 1a-5 to 8, 1a-10 to 17 were then obtained by preparative thin layer chromatography (5%
MeOH/CH2Cl2).
2-((3r,5r,7r)-Adamantan-1-yl)-N-(2-oxo-2-(4-(2-((4-sulfamoylbenzyl)oxy)benzyl)piperazin-
1
1-yl)ethyl)acetamide (1a-5). Yield: 54.0%. H NMR (400 MHz, Chloroformꢀd, ppm) δ: 7.95 (d,
J = 8.4 Hz, 2H), 7.58 (d, J = 8.4 Hz, 2H), 7.34 (dd, J1 = 7.4 Hz, J2 = 1.7 Hz, 1H), 7.27 ꢀ 7.23 (m,
3H), 6.99 (td, J1 = 7.4 Hz, J2 = 1.0 Hz, 1H), 6.92(d, J = 8.2 Hz, 1H), 6.49 (t, J = 3.8 Hz, 1H),
5.16 (s, 2H), 4.02 (d, J = 4.0 Hz, 2H), 3.63 ꢀ 3.59 (m, 4H), 3.36 (t, J = 4.9 Hz, 2H), 2.47 ꢀ 2.42
(m, 4H), 2.01 (s, 2H), 1.96 (s, 3H), 1.70ꢀ 1.60 (m, 12H). MS: m/z 595.30 [M + H]+, calc’d for
C32H43N4O5S: 595.29.
ꢀ 29 ꢀ
ACS Paragon Plus Environment